膝關節角度量測裝置開發與裝置阻力對下樓梯影響之探討

Abstract

走樓梯是生活中常見的功能性活動，相對於平地行走是相當劇烈的活動。而跌倒為年長者十大死因之一，其中下樓梯因重心前傾，跌倒風險甚高，且下樓梯現有對策難以實時監控樓梯行走狀況。 本研究旨在開發一種膝關節角度監測穿戴式裝置，並透過增加彈力帶來分析此外力如何影響下樓梯時的動態平衡。我們將慣性感測元件與彎曲感測器結合於軟性膝蓋穿戴輔具中，使其能夠在下樓梯的過程中實時感測使用者的動作，實驗結果顯示，增加彈力帶於該裝置能增加質心移動速度，進而降低使用者在步態循環中處在動態穩定的時期。為了增強裝置的功能，我們結合機器學習模型來預測使用者的膝關節角度，以提供較準確的角度評估，在調整模型參數後的驗證結果顯示，角度預測誤差值為最低可達10.8度。

Stair negotiation is a common functional activity in daily life, and it’s also an intense exercise compared to level-walking. Falling is one of the top ten causes of death among the elderly. The risk of falling during stair descent is very high due to forward tilt of body weight. This study aims to develop a wearable device for monitoring knee joint angles and to analyze how the addition of elastic bands influences dynamic stability during stair descent. We integrated inertial measurement units (IMUs) and flex sensors into a soft wearable knee brace, enabling real-time monitoring of the user's movements while descending stairs. Experimental results indicate that adding elastic bands to the device increases the speed of the center of mass movement, thereby reducing the time spent in dynamically stable phases of the gait cycle. To enhance the functionality of the device, we incorporated machine learning models to predict the user's knee joint angles, providing more accurate angle assessments. After adjusting the model parameters, the validation results showed that the angle prediction error could be reduced to as low as 10.8 degrees.